Reverse Engineering a Profitable System (REPS) | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Very Profitable System - Roi system 355 replies

Reverse engineering order flow from naked chart 18 replies

MA, Gann and Ichimoku system (profitable system!) 60 replies

Fractal Reverse Engineering Journal 21 replies

METATRADER REVERSE ENGINEERING 3 replies

Attachments: Reverse Engineering a Profitable System (REPS)
Exit Attachments

Reverse Engineering a Profitable System (REPS)

  • #1
  • First Post: Edited Nov 22, 2015 4:52pm Aug 7, 2015 12:48pm | Edited Nov 22, 2015 4:52pm
================================================================================================
Major update to thread on post #297 - http://www.forexfactory.com/showthre...74#post8604674
================================================================================================

Major Edit - 8/11/15, 8/18/15

REPS trading rules -
We've come to the point where we can put our rules to "pen and paper". We may find it necessary to tweak them as we go but I'm trading them as listed below in the attached explorer.

This is a long term trend trading system with an emphasis on low risk, high return.

Instruments - Just about anything that moves, but we may select pairs that tend to trend regularly (think Eur/Aud vs. Eur/Gbp)

Trading Criteria -

1. Initial series entry
Break of the 20 period Donchian Channel on the Weekly chart would constitute the beginning of a trend. Initial entry taken at that point.
Position size at trader's discretion but the goal here is to make this a low risk, or in other words not a big hit to our trading balance. So I'm sticking
with a low % of account balance. .25 - 2% would seem a valid range.
Stoploss on initial entry (and all subsequent entries) will be the opposing side of the 20 period Donchian Channel on the H4 tf.

2. Trade management
We want to get these trades to BE as soon as practicable. There's always going to be a trade off between moving to BE to preserve our trading capital
vs. risking loosing a position by stopping out at BE. Look for market structure and price movement to help determine a good spot but should be
around 50 pips depending on pair.
EDIT 8/18/15 - Trail s/l on the H4 DC until BE

3. Add on trades
New trades will be added on a break of the 20 period Donchian Channel on the Daily tf. Obviously not every break is taken. Location of trade will be
determined when trade will allow for use of MM as discussed below.
EDIT 8/18/15 Aggressive addon - If H4 DC pushes in opposite direction of Daily DC then reverses in direction of trend, then take entry on break of H4 DC with s/l at opposite H4 DC as always.
If we are attempting to enter a trend that is already established we can enter on either of the above 2 criteria if conditions are met.

4. Money management
Once our initial series entry is at BE and risk free we will use the floating profits from our position(s) to fund further positions. We will lock in via moving s/l
50% of our floating profit and use the other 50% for our next entry. For example, if we risk $10 on initial entry and then we have $30 of floating profit
at time of addon trade, we would "lock in" $15 of profit and use $15 to open next position. Typically this would increase our position size each step of
the way, while adding no further risk.
EDIT 8/18/15 - If a trade is taken out at a loss then we only reenter using a MM reset to the lesser of the newly calculated entry stake minus the amount lost in previous trade OR our initial entry risk %...whichever is greater. (Never put locked in "50%" at further risk in financing addons)

5. Series exit
When price hits the opposing 20 period Donchian Channel on the Daily tf this will signify the end of the trend and all positions will be closed.

================================================================================================

ORIGINAL MATERIAL -


Going into the weekend on a "Dog Days of August Trading" Friday afternoon, I thought it might be interesting to develop a new...or at least revised...trading system with the input of the FF community.

It seems typically as we develop a system to trade we look at charts and indicators and how they may tell us where the market is going and how we can enter. As the system evolves (or oftentimes devolves) we make changes and corrections as we are not arriving at our desired outcome in one way or another.

So instead, I'd like to look at it the other way around, so to speak. We'll start with where we want to end up and see if we can't develop a way to get there...

To start with, just a little bit about my trading style.

I favor long term trend following. Ultimately I would like a system that's low maintenance (in terms of computer time), stress free (which means, among other things, low risk) and of course wildly profitable...

Don't get me wrong...I like to dabble in shorter term trading, higher risk/leverage ideas, etc. etc. But that's mainly a past time or trading fix solution. For a go to, make a living system, those just won't work for me.


Everyone's welcome to chime in and participate and share ideas. That doesn't mean I'll necessarily incorporate them in what I'm doing, or in this thread, but I fully expect to get some good ideas from you guys as hopefully I share a few with you.

I just ask that everyone play nice and get along...
So....off we go
  • #2
  • Aug 7, 2015 1:18pm Aug 7, 2015 1:18pm
So, for our first hurdle...or is it our "last" hurdle... Where do we want to end up.


Take a look at these 3 charts
Attached Image (click to enlarge)
Click to Enlarge Name: GBPAUDiDaily.png Size: 62 KB

Attached Image (click to enlarge)
Click to Enlarge Name: EURNZDiDaily.png Size: 79 KB

Attached Image (click to enlarge)
Click to Enlarge Name: EURUSDiDaily.png Size: 68 KB


These represent significant trends that have taken place recently. Take a look at many pair charts on a long term basis and you'll see these long tail trends.
Now there has been, and continues to be, much discussion and even argument on FF and elsewhere as to whether this type of price action is random, quantifiable, manipulated by power brokers, etc. etc.

Frankly... I DON'T CARE. The important point for me is they do in fact happen on a regular basis. And in my opinion, they are...or at least should be...tradeable.

So again, our first (last) goal is to get into these long term trends as soon as possible, and then trade the heck out of em on the way up/down and then get out with as much of the trend as possible.
  • #3
  • Aug 7, 2015 1:58pm Aug 7, 2015 1:58pm
When it's trending like that, I use Donchain Channel, set to 20 on higher timeframes, and 55 on lower timeframes, with 200 MA. That's it. You can either add to your position or take profit as it bounces around in the channel. Kind of like the Turtle Traders method, except not as many channels.
  • #4
  • Aug 7, 2015 2:48pm Aug 7, 2015 2:48pm
Quoting MisterWhite
When it's trending like that, I use Donchain Channel, set to 20 on higher timeframes, and 55 on lower timeframes, with 200 MA. That's it. You can either add to your position or take profit as it bounces around in the channel. Kind of like the Turtle Traders method, except not as many channels.

MW...Thanks for your input.
Donchian channels certainly have potential here...we'll look at that in a minute.

Our first task is to look at these 3 examples.
With the benefit of hindsight, what criteria would have kept us in these pairs for the bulk of the trend?

For the purposes of our discussion, we'll term a series of trades in a given pair as...well...a series.

So to determine how long we kept our series in this trend we have to first determine what would constitute our exit.

We'll eventually be talking about 2 different types of exit. One will be in relation to our Money/Trade management and we'll discuss that later. But first we can look at exiting a series due to the trend (as defined by our criteria) being over, at least for the time being.

So let's talk about possible exits for our 3 examples that would have kept us in these trends.

MW mentioned the Donchian Channel with a setting of 20. Let's take a look.

Attached Image (click to enlarge)
Click to Enlarge Name: Traders Way MetaTrader 4.png Size: 135 KB


So if we set up the criteria of exiting our series at the break of the Donchian channel we see that the E/N and G/U would still be active and ongoing.
The E/U would have exited on the 28th of April after having been kicked out toward the beginning of the trend on the 30th of June, 2014.
So we would have potentially captured about 2,500 of those 3,300 pips, depending on our initial entry. Not too bad.

Next we'll look at Moving Averages.

Any more ideas???
  • #5
  • Edited 3:56pm Aug 7, 2015 3:22pm | Edited 3:56pm
Quoting spyderman
{quote} With the benefit of hindsight, what criteria would have kept us in these pairs for the bulk of the trend?
Balls.

If you really have a pair, you would add to the position at confirmation points (where the price touches either the upper or lower channel boundary, depending on if the trend is bullish or bearish). Add to the bottom bounce, if bullish; add to the upper bounce, if bearish. You could also take profit at these points, re-establishing a fresh position at the bounce.

Actually, the real criteria are: if price breaks significantly above/below the channel boundary, or if price breaks above/below the 200 MA, you would exit the position. You could throw a stochastic in there, set to: 15,5,5 for further confirmation. Your initial entry would be after a couple of confirmation bounces, to avoid getting kicked out at the beginning, so to speak. You don't make the entire move, but enough.

THIS ONLY WORKS ON CLEARLY DEFINED TRENDS. As someone mentioned in another thread, trends like this are becoming more rare, in modern markets. This is an old system and does not usually work. Typically, the price goes sideways or follows an erratic pattern.
  • #6
  • Aug 7, 2015 4:07pm Aug 7, 2015 4:07pm
Of course, you never know what will happen on the right edge, but here is my analysis - nothing new.
The first problem is identifying the trend change, then staying in the trend with possible additions to the position and finally exiting.
I'm attaching 2 charts which make up the entire trend on the EU. I think my notes show my method for the above stages.
The EMAs are 40 and 15 with the 40 changing colour with slope. The first entry occurs when the first full candle is in the direction of the 40EMA and below the 15 EMA.
Attached Image(s) (click to enlarge)
Click to Enlarge Name: Long Trend.png Size: 37 KB
Click to Enlarge Name: Long Trend002.png Size: 41 KB
  • #7
  • Aug 7, 2015 4:18pm Aug 7, 2015 4:18pm
Quoting MisterWhite
{quote} Balls. If you really have a pair, you would add to the position at confirmation points (where the price touches either the upper or lower channel boundary, depending on if the trend is bullish or bearish). Add to the bottom bounce, if bullish; add to the upper bounce, if bearish. You could also take profit at these points, re-establishing a fresh position at the bounce. Actually, the real criteria are: if price breaks significantly above/below the channel boundary, or if price breaks above/below the 200 MA, you would exit the position. You...

Whoaaaaaa...getting ahead of ourselves. Lol, but thanks for the input. We'll discuss entries in due time. We're just establishing a series exit criteria at the moment.
  • #8
  • Aug 7, 2015 4:29pm Aug 7, 2015 4:29pm
this can be answer
Attached Image (click to enlarge)
Click to Enlarge Name: answer.jpg Size: 153 KB
  • #9
  • Aug 7, 2015 4:37pm Aug 7, 2015 4:37pm
Quoting profitkeeper
Of course, you never know what will happen on the right edge, but here is my analysis - nothing new. The first problem is identifying the trend change, then staying in the trend with possible additions to the position and finally exiting. I'm attaching 2 charts which make up the entire trend on the EU. I think my notes show my method for the above stages. The EMAs are 40 and 15 with the 40 changing colour with slope. The first entry occurs when the first full candle is in the direction of the 40EMA and below the 15 EMA. {image} {image}


Thanks for contributing PK.

Let's look at some MA's.

Here's our 3 examples with a 20 (red), 50 (blue) and a 100 (green) MA added.

Attached Image (click to enlarge)
Click to Enlarge Name: Traders Way MetaTrader 4A.png Size: 191 KB


If we're thinking of exiting on an MA cross, you'll see that on the E/N and G/U all but the 20 would have us still in the trend currently, just as was the case with the Donchian Channel. So we haven't gained any new information.
On the E/U we see the 20 exiting frequently as we might expect and the 50 is kicking us out a couple of times. The 100 had us out on 4/28/14 as the DC did but keeps us in all the way down. We get a later exit on the 100 than the DC which cost us a couple hundred more pips on the later exit.

So far the Donchian is giving us a better exit on these examples
On the E
  • #10
  • Aug 7, 2015 4:38pm Aug 7, 2015 4:38pm
Quoting fulltime
this can be answer {image}

FT. Thanks for stopping by. Not sure where you're heading with this one. Perhaps you could expand on it?
  • #11
  • Aug 7, 2015 4:44pm Aug 7, 2015 4:44pm
Let me take a moment to emphasize my goal here.

I'm looking at some recent trends that would have been very profitable as we shall find out.
Based on those examples (E/U, E/N, G/U) how could we engineer a system that would have traded those trends with as much profit as we could.

At this point the focus isn't on what may or may not happen down the road or on future trends.

If we can maximize what could have been made on what did happen, I am confident that same system can be used in the future on what will happen.

As the thread unfolds, if it in fact does, then we shall explore future trends using our REPS system.
  • #12
  • Aug 7, 2015 4:55pm Aug 7, 2015 4:55pm
if a line respected 3 or more times in long term (4h and over). this is a good oppurtunity. Because big boys respect these lines. How to play is up to you .
Attached Image (click to enlarge)
Click to Enlarge Name: answer2.jpg Size: 204 KB
  • #13
  • Aug 7, 2015 4:57pm Aug 7, 2015 4:57pm
Quoting fulltime
if a line respected 3 or more times in long term (4h and over). this is a good oppurtunity. Because big boys respect these lines. How to play is up to you . {image}
So what are you suggesting as the exit...and how might that apply to our other examples?
  • #14
  • Aug 7, 2015 5:00pm Aug 7, 2015 5:00pm
Well the thread has been moved to Trading Discussion from Trading Systems. So be it.

There is a system here, I'm just going about it a bit differently. Ah well...we shall soldier on.
  • #15
  • Aug 7, 2015 5:11pm Aug 7, 2015 5:11pm
When you find a line(in weekly monthly) you know that pair can turn from that line.See the move than enter .That move will be few hundred to thousand pips. but its boring waiting days weeks etc. or trade in 4h lines wait for one two days... So on...
  • #16
  • Aug 7, 2015 5:33pm Aug 7, 2015 5:33pm
So let's move forward with the Donchian Channel for the moment for our series exit. If other ideas come up later we'll look at those.

So to reiterate, we will exit all trades in a series when the opposite side Donchian channel on the daily tf is hit. Or more accurately we can move our group stop down as the DC changes.

Having decided on a series exit, we must next decide how we would like to make our initial entry on the series.
Now we can continue to use the DC, entering the opposite direction on the previous series exit. Or we can establish additional filter criteria.
Some possible options are additional indicators, price action, higher TF analysis, etc.

Let's consider.
  • #17
  • Aug 7, 2015 6:17pm Aug 7, 2015 6:17pm
For the moment let's concentrate on the E/U example as it's the longest trend. Then we can apply our findings to the other.

So first let's look at consulting a higher TF for clues on our initial entry.

Looking here at the Weekly E/U with our same Donchian Channel we see the trend changed down on 7/18/14 at a price of 1.3503
We can see the entry point as we drill down to the Daily
Note this is after the DC exit that we mentioned in the above charts.

Attached Image (click to enlarge)
Click to Enlarge Name: EURUSDiWeekly.png Size: 50 KB

Attached Image (click to enlarge)
Click to Enlarge Name: EURUSDiDaily.png Size: 50 KB
  • #18
  • Aug 7, 2015 6:29pm Aug 7, 2015 6:29pm
So for reference purposes going forward, let's identify the different segments discussed as we go. That way we can go back and modify a specific segment if we see the need.

So thus far we have...

Segment 1 - Trade series exit.
Segment 2 - Trade series initial entry.
  • #19
  • Aug 7, 2015 7:05pm Aug 7, 2015 7:05pm
Quoting fulltime
When you find a line(in weekly monthly) you know that pair can turn from that line.See the move than enter .That move will be few hundred to thousand pips. but its boring waiting days weeks etc. or trade in 4h lines wait for one two days... So on...
You can try to trade news, in-between waiting.... and probably lose what you gained waiting for your trend trades to play out.
  • #20
  • Aug 7, 2015 7:49pm Aug 7, 2015 7:49pm
Okay, since we've addressed Segment 2 - Initial series entry, we need to also talk about

Segment 3 - Money Management

There will be many aspects to MM that we will address as we go but for now we have to think about initial position sizing.

As mentioned at the outset, I want this to be a low risk trading system. Low risk can mean a lot of different things to different people.
To a large extent it will depend on you as a trader, your resources, risk tolerance, etc. There's umpteen threads about that.
For the sake of simplicity let us begin with the assumption that we will risk 1% of our trading capital for an initial series entry.
We can and will no doubt adjust this (I suspect lower) as we go forward, but, so be it for now.

We will figure the initial balance in our trading account is $10,000.
Trading the Eur/Usd our value per pip is $10 for a full standard lot, $1 for .1 lots and 10 cents for .01 lots
So we are wanting to risk 1% or $100

So here's a representation of our initial entry point on the weekly chart with the entry candle on the far right.

Attached Image (click to enlarge)
Click to Enlarge Name: EURUSDiWeekly.png Size: 62 KB


We know our sell stop level, the lower donchian channel. We know we want to risk $100 for this initial trade, so to determine our position size we need to know what our initial stoploss will be.

Again we have choices. We can use -

Price action/structure
ADR
Set stop
Other side of Donchian Channel on Daily/Weekly

Anyone have any other suggestions.
0 traders viewing now